Development of Business Intelligence Outlier Detection and Predictive Analytics System in Managing Occupational Fraud and Loan Defaults in Banks.

Project reference number: GSBS-2019-111

Banks have faced significant regulatory requirements since the financial crisis. These new regulations have come in a period of slow economic growth and historically low-interest rates environment which have limited revenue opportunities for them while increasing compliance costs. Increased capital and liquidity requirements have further reduced returns and force banks to cut costs including risk management costs. At the same time, risk management within the banking sector has experienced significant evolutional changes in the years since the financial crisis as a result of current market complexities and high incidences of frauds and loan defaults. The realities of the current operating environment provide strong incentives for banks particularly in developing economies to fundamentally rethink their risk management approaches.

The volume and type of ‘Big Data’ that banks are currently collecting and warehousing has positioned them to exploit advanced analytics that can reduce costs while providing insights into current and emerging fraud-related risks and guarding against loan defaults. Big data analytics can provide deeper insight into the interactions of these risks and causal factors. The ability to harness larger and more diverse data pools is critical to both reducing occupational fraud-related losses and increasing revenue by highlighting business opportunities.

The Project
The aim of this project is to develop a Business Intelligence Outlier Detection and Predictive Analytics System which aims to learn from historical transaction information in order to retrieve patterns that allow differentiating between normal and abnormal activities. The model will be developed to distinguish between legitimate and fraudulent transactions in real-time and contact customers to verify suspicious transactions. The predictive analytics model can help in predicting loan defaults by monitoring borrower’s behaviour to anticipate and respond to default risk. This model will combine existing historical transactional, demographic and product data with behavioural information, from both internal and external sources to generate a robust data set that can be used to detect, predict and mitigate risks efficiently across multiple banking domains.

Informal inquiries can be made to Dr Nwafor, Chioma ([Email Address Removed]) with a copy of your curriculum vitae and cover letter.

Supervisor Research Profiles
Director of Studies: Dr Chioma Nwafor
GCU Research Online URL:
http://researchonline.gcu.ac.uk/portal/en/persons/chioma-nwafor(14a4e6f6-f69d-4057-ba83-0fd6967bc433).html

2nd Supervisor: Dr Mauricio Silva
GCU Research Online URL: http://researchonline.gcu.ac.uk/portal/en/persons/mauricio-silva(f24dd902-7d57-4dce-be32-70b7656ace0f).html
How to apply

The project is available as a 3 years full-time or 6 years part-time PhD study programme with an expected start date of the 1st October 2020.

Candidates are encouraged to contact the research supervisor for the project before applying.

To apply for this project, use the following link to access the online application form, as well as further information on how to apply: https://www.gcu.ac.uk/research/postgraduateresearchstudy/applicationprocess/.

Please send any other enquiries regarding your application to [Email Address Removed]